Doctors' and nurses' perceptions of ethical ... - Wiley Online Library

1PHILOSOPHICAL AND ETHICAL ISSUES

Doctors' and nurses' perceptions of ethical problems in end-of-life decisions Kathleen Oberle RN PhD Associate Professor, Faculty of Nursing, University of Calgary, Calgary, Alberta, Canada

and Dorothy Hughes RN PhD Associate Professor, Faculty of Nursing, University of Calgary, Calgary, Alberta, Canada

Submitted for publication 16 February 2000 Accepted for publication 25 October 2000

Correspondence: Kathleen Oberle, Faculty of Nursing, University of Calgary, 2500 University Drive NW, Calgary, Alberta T0L 0W0, Canada.

Journal of Advanced Nursing 33(6), 707±715 Doctors' and nurses' perceptions of ethical problems in end-of-life decisions Aims. To identify and compare doctors' and nurses' perceptions of ethical problems. Rationale. Ethical problems are a source of tension for health professionals. Misunderstandings or con¯icts may result from differing perceptions of ethical problems. If true collaboration is to be achieved, it is important to understand the perspectives of others, particularly when dif®cult end-of-life decisions must be made. Methods. In this qualitative study a total of seven doctors and 14 nurses working in acute care adult medical-surgical areas, including intensive care, were asked to describe ethical problems that they frequently encounter in practice. Interviews were taped and transcribed. Thematic analysis followed. Results. All participants experienced ethical problems around decision making at the end of life. The core problem for both doctors and nurses was witnessing suffering, which engendered a moral obligation to reduce that suffering. Uncertainty about the best course of action for the patient and family was a source of moral distress. Competing values, hierarchical processes, scarce resources, and communication emerged as common themes. The key difference between the groups was that doctors are responsible for making decisions and nurses must live with these decisions. Each group, therefore, asked different questions when encountering and interpreting sources of moral distress. Conclusions. It was concluded that observed differences between doctors and nurses were a function of the professional role played by each rather than differences in ethical reasoning or moral motivation. Although this was a small qualitative study on one institution, and may not be generalizable, results suggest that doctors and nurses need to engage in moral discourse to understand and support the ethical burden carried by the other. Administrators should provide opportunities for discourse to help staff reduce moral distress and generate creative strategies for dealing with this.

OBERLE K. & HUGHES D. (2001)

Keywords: ethics, nursing ethics, medical ethics, end-of-life decision making

3 Introduction Effective collaboration among doctors and nurses is essential for high quality. For true collaboration to be achieved, it is important to understand the perspectives of others, particÓ 2001 Blackwell Science Ltd

ularly in situations that may be ethically problematic (Pike 1991, Johanson 1994, van der Arend & Remmers-van den Hurk 1999, Sundin-Huard & Fahy 1999, Wurzbach 1999). Research suggests that doctors and nurses may perceive ethical problems differently, and may employ different 707

K. Oberle and D. Hughes

reasoning and decision-making frameworks (GrundsteinAmado 1992, 1993). If health professionals see themselves as having differing moral commitments to a patient, there may be potential for con¯ict. Therefore, the purpose of our study was to explore doctors' and nurses' perceptions of ethical problems in an acute care setting, with the objective of illuminating similarities and differences in ethical reasoning. Signi®cance of this study lies in the fact that enhanced awareness and increased understanding can be important factors in improving collaborative practice (Larson 1999).

Background Support for the idea that nurses and doctors might think differently about ethical issues is lodged in the debate about whether bioethics, the basis of medical ethics, is an appropriate framework for nursing ethics. Bioethics, as ®rst described by Beauchamp and Childress (1989), involved the application of four basic principles: autonomy, bene®cence, nonmale®cence and justice. Thus, bioethics was `the framing of problems and solutions by a relatively small set of concepts: rights, duties, obligations, competence, and justice' (Gustafson 1990, p. 127). Fry (1989) drew a distinction between nursing and medical ethics, suggesting that the value foundations of nursing ethics are derived from the nature of the nurse±patient relationship instead of from models of patient good, rights-based notions of autonomy, or the social contract of professional practice as articulated in prominent theories of medical ethics (p. 9).

Building on this argument, Twomey (1989) asserted that nursing ethics is distinct from bioethics, and that the essential difference could be identi®ed through a metaethical analysis of the de®nition of good. In nursing, this `good' was seen to be rooted in the caring relationship between patient and nurse. Thus the caring that is central to nursing may involve a different type of moral commitment, and hence a different kind of moral reasoning than is represented in bioethics (Parker 1990, Gadow 1999). However, a de®nitive theory for ethical decision making in nursing, as distinct from traditional bioethics, has not been presented (Lipp 1998). Researchers have studied differences in ethical reasoning between doctors and nurses, with varying results. GrundsteinAmado (1992, 1993) interviewed doctors and nurses in acute and long-term care and found that both emphasized the obligation to inform, but nurses placed greatest emphasis on patient dignity, comfort, and wishes, while doctors were most concerned with patients' rights and quality of life. Interviews with Swedish physicians and nurses (Uden et al. 1992) suggested different moral orientations between the two 708

groups, but in a follow-up study their beliefs and values were similar. The authors concluded that the ®rst interview re¯ected participants' professional experiences, while the second interview disclosed their personal experiences as human beings (Lindseth et al. 1994). In another Swedish study (Soderberg & Norberg 1993) nurses were most frequently concerned with problems of relationship and choice of action, in equal proportions, while doctors were most concerned with choice of action. Similarly, doctors and nurses working in acute care in the United States of America (USA) identi®ed different ethical problems in the same cases; doctors focused on quality of life, inappropriate admission to hospital, and cost of care, while nurses were more concerned with patient and family preferences, pain management, implementing treatments, and discharge planning (Walker et al. 1991). Viney (1996) studied the experiences of doctors and nurses in decisions to withdraw treatment and found differences related to roles; doctors were primarily decisionmakers, whereas nurses were information brokers. As a result, doctors suffered moral dissonance while nurses suffered moral distress. Are the differences observed in the above studies a result of variations in ethical reasoning, moral commitments or contextual factors? Researchers have attempted to answer this question, with equivocal results. For example, it has been suggested that orientations to ethical reasoning and moral commitments may be classi®ed as `justice' (an approach based on application of abstract principles of justice or fairness) or `care' (an approach focused on relationships) (Cooper 1988, Parker 1990, Grundstein-Amado 1992, Peter & Gallop 1994). Kuhse et al. (1997) used hypothetical dilemmas to explore the hypothesis that nurses would have a stronger `care' orientation, and doctors a stronger `justice' orientation, and found no signi®cant differences between groups. Conversely, Robertson (1996), using ethnographic methods, concluded that doctors and nurses used the same ethical principles to guide their work, but nurses were more likely to base their interpretation of principles on a virtue/ relationship model, whereas physicians were more inclined to employ a utility-based model. Thus, differences were seen in the kinds of reasoning applied and the kinds of moral commitments held by nursing and medicine. Several studies addressing ethical decision making in nursing (Lipp 1998, Kelly 1998a, b) have demonstrated that environmental or contextual factors are key to nurses' experience of moral distress. However, the question of moral commitments of nurses, as distinct from doctors, was not addressed in these studies. Thus, research indicates that important differences exist in doctors' and nurses' perceptions of ethical problems in

Ó 2001 Blackwell Science Ltd, Journal of Advanced Nursing, 33(6), 707±715

1 Philosophical and ethical issues

practice, and that such differences may become a source of con¯ict (McClure 1991, Johnson et al. 1995, Taylor 1995, Rodney 1998). All authors cited above suggested that further exploration is warranted for theory development and to promote dialogue and foster understanding across disciplines.

Method A qualitative descriptive approach, based on the grounded theory methodology of Strauss and Corbin (1998) was used to determine the types of ethical problems perceived by doctors and nurses, and to compare these perceptions with recent research and conceptual literature about ethical frameworks in medicine and nursing. Ethical approval was obtained from the Research Ethics Board prior to inception of the study. Participants were 14 nurses and 7 doctors working on adult medical±surgical units in one large acute care Canadian hospital. All the nurses were female and all but one of the doctors were male. Nursing experience ranged from 5 to 28 years while the doctors had been practicing up to 20 years. Three nurses and one doctor had some formal ethics education. A nominated sampling technique was used, with nursing unit managers identifying staff nurses whom they believed to be particularly thoughtful about ethical questions. For physician recruitment, a doctor collaborating in the study approached potential participants to consider taking part in the study. Once permission was obtained, participants were contacted by the researchers and the study explained in depth. A time was established for the interview, and informed consent obtained. One-on-one unstructured interviews, ranging in length from 45 to 90 minutes, were conducted at a site of the participant's choosing. Interviews were tape recorded and transcribed for later analysis. All interviews began with the grand tour question, `Please describe for me a frequently recurring ethical problem that you have experienced in practice ± something that has been a common problem for you.' Probes were used to encourage individuals to elaborate on their perceptions, for example, `Why did that raise ethical issues for you? What was the nature of the underlying concern?' Once each interview was transcribed, thematic analysis was performed, based on the methods outlined by Strauss and Corbin (1998). Coding began after both investigators read each transcript several times to get a sense of the whole. Using a constant comparison method, individual meaning units were sorted and resorted as categories, and then patterns, began to emerge. Patterns were combined and consolidated into common themes. Trustworthiness was addressed by

Ethical problems in end-of-life decisions

leaving a clear audit trail in the form of ®eld notes, logs, transcribed tape recordings and taped interviews. Because two researchers were involved in the data collection and analysis process, critical dialogue and confederate review supported dependability. Analysis was conducted in three phases: (i) doctors' and nurses' interviews were analysed separately and themes identi®ed; (ii) overarching themes common to both were determined and (iii) similarities and differences within each overarching theme were noted.

Results In the initial analysis, differences between doctors' and nurses' ethical concerns appeared extensive. However, with repeated review of the transcripts, similarities in underlying issues became evident. Stories of particular patient and family centred situations were often used to illuminate ethical problems encountered in the participants' clinical practice. The focus of this paper will be end-of-life decision making, which was the most frequently identi®ed ethical problem in both groups. Differences between doctors' and nurses' ethical concerns were primarily related to their perceived mandates as caregivers. Perspectives on ethical problems appeared different because doctors bore the burden of having to make the decisions and write the orders, whereas nurses' burden entailed living with the decisions made by someone else. Doctors experienced true ethical dilemmas as they were required to choose between two mutually exclusive courses of action (for example, to write or not to write a `do not resuscitate' order), and they reported agonizing over whether they had made the `right' decision. Both doctors and nurses were highly concerned with the decision-making process, however, because the decisions were not theirs to make, nurses' concerns centred on how and why others came to the decision. Thus, while doctors' experiences around end-of-life decisions represented moral dilemmas, nurses' experiences were primarily associated with moral distress or what Pike (1991) has termed `moral outrage'. Misunderstandings and even con¯icts with doctors, resulting from an inability to in¯uence decisions and decision-making processes were sources of moral distress for every nurse interviewed. In contrast, none of the doctors interviewed identi®ed con¯icts with nurses as a source of moral distress. They acknowledged that nurses sometimes had problems with care decisions but did not de®ne these as ethical problems. In essence, doctors questioned themselves, and nurses questioned doctors. Despite these differences it became evident, as patterns and themes converged, that the core problem was essentially the


709


same for doctors and nurses; what was at variance was what they were expected to do about it, and how it played out in practice. The core problem was conceptualized as witnessing suffering of patients and families. The moral obligation of both doctors and nurses was to minimize that suffering. Traditional ethical principles of bene®cence, autonomy, and justice were much in evidence in the participants' thinking, but concerns were seldom framed in those terms. Instead, participants talked about how they de®ned `good' in any given situation and how various contextual features impacted on their ability to respond to the suffering. These contextual features, all linked to uncertainty in patients' prognoses, are described under the themes of (i) competing values; (ii) scarce resources and (iii) hierarchical processes. Communication emerged as a distinct theme that wove throughout all the contextual themes. Themes were not mutually exclusive or independent, but rather overlapping and intertwined. For the sake of discussion, each will be discussed separately.

The de®ning feature: uncertainty In end-of-life decision making, uncertainty about probable outcomes was the de®ning feature, leading to considerable deliberation and re¯ection about the `right thing to do'. At what point did patient suffering outweigh the probability of a positive outcome, and at what point should treatment be stopped? Even in the so-called futile cases there remained the possibility, however slight, that a positive outcome might result from further treatment. Indeed, some patients were known to defy the odds and live to achieve a reasonable quality of life. However, treatment that merely prolonged death and caused needless suffering was a source of great distress for those committed to reducing that suffering. The quandary in which doctors found themselves was stopping treatment. Although often referred to as the family's decision, it was the doctor who had to take ®nal responsibility. Doctors were expected to make the `right' decision, even when right was by no means clear. As one doctor expressed it, We're trained to make a decision and do it ¼ But yet if you take the whole thing down the line it might not be the right thing to do. There's always the unknown.

By contrast, nurses experienced moral distress in these uncertain situations when they believed that the wrong course of action was being followed and they were contributing to the patient's misery. Nurses tended to judge the doctors' actions as a function of whether or not they had made the right decision and written the right order. Nurses often had clear ideas about what was acceptable and what 710

was not, re¯ecting a degree of certainty not evident in the doctors' responses. Distress arose for nurses when they perceived that the patient's suffering was intensi®ed because doctors just could not or would not write the àppropriate' orders. There was, however, paradoxical thinking evident in many of the nurses' responses. Whereas they expressed concern that the doctor had not made the `right' decision, they also recognized that the decisions were dif®cult, and that outcomes were, indeed, uncertain. One doctor captured the source of the difference. His views appeared to mirror those of the nurses. He noted, I sense a lot of time the nurses [wonder] what I'm doing to these patients. Until you're in that position of actually having to make the ®nal decision to turn somebody off, it's a lot easier to say it and a lot harder to do it.

Contextual features Competing values Decisions about care and treatment at end-of-life are intrinsically value-laden, and recognition of this fact raised concerns for all participants. Questions arose about whose values should carry the most moral weight in any given situation, and whose values had, or should have had, the most impact on decisions taken. This was particularly problematic if the patient was unable to speak for self. The value embraced by all participants was the good of the patient. In the absence of a patient voice, patient autonomy was not a realistic goal, but doing `what the patient would want' was a reasonable alternative. Unfortunately, it was seldom clear just what the patient would want, and the parties to discussion about continuing or discontinuing treatment sometimes possessed different beliefs about the patient's wishes. When the patient was without voice, it was generally expected that the family would guide the decisions, acting in the best interests of the patient. However, concerns were expressed that families did not always appreciate the implications of continued treatment and sometimes acted in their own best interests rather than those of the patient. Although it was acknowledged that the family had legitimate needs, participants felt those needs should not be permitted to extend the patient's suffering. Doctors wanted the patient or family to have input, but the perception that patients and families might not understand the realities of the situation was a problem. Doctors reported agonizing over what to tell the family and whether the ultimate decision was swayed too much by their own values. As one doctor explained, You often ®nd yourself putting a spin on the facts that might not be completely objective. There is a tendency not so much to exaggerate


1 Philosophical and ethical issues as to emphasize the facts that are in favour of your opinion ¼ maybe I'm misrepresenting it to the patient, you know, using my value system rather than his.

In contrast, nurses' concerns about con¯icting values were focused primarily on why, in their perception, the doctor was not acting in the patients' or the families' best interests. Nurses thought doctors too often acted on their own values, rather than those of the patients and families. This translated into failure to talk to families in a timely manner, to provide adequate explanations, or to write DNR orders. They suggested that doctors often refused to think about the suffering they caused. One nurse described the problem: Of course people can be kept alive, but I think physicians have to step back and say, `what are we doing here?' ¼ the physicians themselves


despite their professional assessment and detailed understanding of the patient's condition. Again, their problem was with their inability to reduce the patient's suffering, which in turn resulted in their own suffering (experienced as moral distress), as illustrated below: [We're] very frustrated. I think we hurt a lot for the patients. It doesn't matter what we tell most of the physicians, about the pain or suffering, ¼ about how miserable they are with all the treatments they're getting, etc. It's almost like it's falling on deaf ears.

Thus, both doctors and nurses experienced an inability to exercise moral agency as a result of the constraints imposed by hierarchical structures. Interestingly, neither group appeared to be aware that hierarchy and a sense of powerlessness were issues for the other.

have to feel comfortable with letting this patient die ¼ it's hard for physicians to let go.

Nurses often perceived that differences in values between doctors and nurses constrained nurses from acting on their own beliefs. Doctors' concerns about patient and family suffering were similar to those of nurses, but doctors were nonetheless responsible for making dif®cult choices, the reasons for which were not always apparent to nurses. Hierarchical processes Doctors experienced problems with the hospital hierarchy in that they did not always feel free to make the best decisions or obtain the best care for their patients. This became problematic, for example, when they wanted to admit a patient to an intensive care unit and other doctors or hospital policy prevented that from happening. The issue was often linked to scarce resources, but was also a question of values as the prevailing opinion might be that the patient was `not worth saving'. Doctors felt constrained in decision-making by hospital administration, other health care providers, and by patient/family expectations. Scenarios involving the right of the family to demand futile treatment emerged as a case in point. As one doctor explained: Now if you look at it ethically, we're not obliged to continue with futile care. I think people would agree with that. But the family complains to administration or the press and almost always hospital administration will not support you.

Thus, hierarchy emerged as a separate theme for doctors because decisions were out of their control. For nurses, ethical problems were related to their `lower' position on the hierarchical structure: not being listened to by doctors; being expected to remain silent even when witnessing wrong choices; being unable to impact on decisions,

Scarce resources Perhaps not surprisingly, availability of resources emerged as a concern, although again the problem presented differently in the two groups. For doctors, the main issues revolved around their responsibility for resource allocation. They were keenly aware that their decisions about one patient might impact on another. For example, because of a shortage of beds, a decision to continue treatment for one patient might mean another patient is denied care. Being cast in the role of gatekeeper caused doctors considerable distress. A related issue was whether patients and families should be permitted, despite cost implications, to demand treatment in futile cases. One doctor suggested that `We've gone so far in the direction of respecting individual autonomy that societal justice has suffered greatly'. For some, this translated into a more general concern about how resource distribution decisions were made at a higher level: If you have an eloquent spokesperson or a strong lay interest group to support a program the resources made available to that program are far greater than those made available to other programs that are either not politically interesting or don't have a strong interest group.

Nurses, too, had resource allocation issues, but the majority of their concerns were related to their inability to provide quality care because of ®nancial constraints and staf®ng cutbacks. One nurse summed up the problem: My standards have lowered ¼ I can't meet needs the way I used to. It's physically and emotionally impossible, on the majority of my shifts, to meet those needs, which leads to a lot of frustration.

Here again, the underlying issue was the same for both groups, but the perspective was different because of different responsibilities.


711


The process: communicating For every participant, communicating was Ethical issues arose for doctors about what because of the uncertainty inherent in the also out of a desire not to do harm. As one communication with patients and families

a key factor. to tell, partly situation, but doctor stated,

¼has to be done with optimism. I hate to take away people's hope because that doesn't leave them anything. You don't want to say, well, that's it for you. Nothing I can do.

Doctors also had concerns that misperceptions of the intent of communications might cause families further distress. Timing was also a potential problem. In one doctor's view Nurses will say they need to know the code status on this ¼[but] some days you just can't talk about it. [You] might have just told the patient he or she has cancer or something really bad might have happened ¼ we have to be sensitive. Sometimes the nurses' time might not be the same as the physician's or the patient's time, so there will be con¯ict.

Nurses' concerns were whether patients and families had accurate and adequate information at the appropriate time, particularly in situations where they were expected to make a choice around continuation of treatment. Nurses did not want to increase the family's burden because `The ®nal choice could affect them for the rest of their lives. Are they really well enough informed to make a decision?' They felt that doctors often did not communicate in a timely manner with families: I feel [doctors] approach families too late, when the families are stressed. And I think they are unable to even approach this problem. They ®nally get around to it ¼ [but] in quite a few of the cases I feel that [families] have gone through a lot of unnecessary suffering.

They also worried about the manner of presentating information to patients and families because, `When the doctor leaves, you'll ask the patient what the doctor said to them. And they have no idea what was discussed with them. No layman's terms are used.' Clearly poor communication was a source of distress. Conversely, effectual communication was offered by both groups as the solution to many of the problems: Ìf doctors would just talk to the family'; Ìf doctors would just listen to us'; Ìf we [doctors] knew how to present our views to the patient without withdrawing hope'.

Discussion It is implied in the nursing literature that the foundational ethic for nursing and medicine are different ± that nurses have 712

a stronger care orientation, while medical ethics are based more on justice (Fry 1989, Parker 1990), often presented as `care vs. cure'. Recent research suggests that this belief has permeated nursing education, and is being adopted by nursing students. Joudrey and Gough (1999) explored attitudes of baccalaureate nursing students and found a prevailing belief that `nurses care and doctors cure' (p. 1157). Participants saw physicians as being focused more on technical, medical considerations, whereas nurses were more holistic. Joudrey and Gough's participants appeared almost contemptuous of the perceived narrow perspective of physicians. The authors noted ¼ a perception that the goal of nursing is altruistic whereas physicians were painted as more motivated by either self-interest or advancing medical science (p. 1158).

Our results did not support those beliefs. Instead, the study revealed striking similarities between doctors and nurses in the kinds of ethical problems they faced. The most frequent presenting problem was the same for both groups: end-of life decision making. The underlying moral issue, that is, the suffering of others, and the obligation to respond to that suffering, was the same for all. Differences did not appear to lie in the nature of the moral response, nor in the kinds of moral reasoning used: all participants appeared to use care as moral motivation. Rather, differences appeared to be related to professional roles, the kinds of responsibilities each group had in the situation, and the resultant questions for which they needed answers. Examples are shown in Table 1. This moral response to suffering generated an incredible sense of moral burden in both groups. There was, however, little recognition by either group of the burden carried by the other: nurses acknowledged that decisions were dif®cult, but they did not appear to recognize the moral implications for doctors. Similarly, doctors were aware that their decisions could cause dif®culties for nurses, but appeared to give it little conscious thought. This ®nding is congruent with previous research ®ndings that ethical con¯icts with nurses were seldom an important concern for doctors (Gramelspacher et al. 1986, Walker et al. 1991). While we did not ®nd a difference in moral orientation as such, we did note some differences between groups. In particular, evidence suggested that nurses felt constrained by doctors from being able to act on their beliefs, whereas doctors appeared to feel no such constraint from nurses. Thus, the hierarchical structure emerged as a key element in nurses' distress, which is congruent with the contention that nurses are not free to act as moral agents in the acute care setting (Yarling & McElmurry 1986, Rodney & Starzomski


1 Philosophical and ethical issues


Table 1 Comparison of doctors' and nurses' questions generating moral distress Questions Doctors

Nurses

Uncertainty What is the prognosis? What is the best decision to minimize suffering in the short and long term?

How was the decision made and was it the best decision? What can I do (if anything) to minimize suffering under this decision?

Competing values Whose values and beliefs should I/did I attend to in making the decision My own? The patient's family's? The institutions? Society's? Other caregivers'? Hierarchical processes Am I free to be a `moral agent' and act on my beliefs? Who will support me if I do? Can I in¯uence anyone else's actions? Should I try to? Scarce resources Would someone else bene®t more from the scarce resources? Do I need to take action to make resources available for another? Communication What should I tell? When should I tell? Was the information biased?

4 1993). By extension, however, one can conclude that the hierarchy was also a feature of the doctors' moral landscape as it de®ned the doctor as decision-maker. Contextual features such as family and bureaucratic demands might also constrain doctors' moral agency. Clearly there is a need for considerably more dialogue about ethical issues across disciplines. If the ®ndings of Joudrey and Gough (1999) are any indication, nurses may come to the care situation with preconceived (and possibly erroneous) notions about the moral commitment of physicians. We found little to support the beliefs of a `cure vs. care' orientation among doctors in our study. Similarly, in a recent Swedish study (Sorlie et al. 2000), women physicians narrated stories related to both àction' and `relation' ethics, which suggests an alignment with the `care' perspective as de®ned in nursing literature (Gadow 1999). Interestingly, physicians in the Swedish study perceived that there were different perceptions across disciplines and emphasized a need for discussion between colleagues about ethically dif®cult care situations. It was interesting to note that participants described the contextual features of ethical situations and their responses

Whose values and beliefs were attended to in making the decision The doctor's? The patient's and family's? The institution's? Society's? Other caregivers'? Am I free to be a `moral agent' and act on my beliefs? Who will support me if I do? Can I in¯uence anyone else's actions? Should I try to? Is this the best use of resources? Will we have enough resources to provide comfort and minimize suffering? What was told? Was it told at the best time? Was the information adequate? Accurate? Biased? What am I free to tell? What should I tell?

to them far more often than they described actual ethical dilemmas. Problems for both groups appeared to rest primarily on the nature of their responses to the patient, and the process by which decisions were made. Much of the nursing and medical ethics literature deals with moral reasoning and the cognitive processes by which one comes to a practice decision, but what was evidenced in our study was not related to cognitive reasoning as such. Instead, the issues were around the moral obligation to respond to another's suffering, and the nature of that response ± a kind of ontological commitment to other. Ray (1998) has suggested that ethical decisions are shaped by moral interactions among patients, families and caregivers, and are guided by deep values about vulnerable human beings. Bishop and Scudder (1996) support a similar view related to nursing ethics. Our results suggest, however, that such a commitment is also present in physicians, and ought not to be considered a source of difference in care situations. In collecting data, one question we had intended to ask, `Was the problem resolved to your satisfaction?' seemed inappropriate, because the problems identi®ed were almost never `resolvable' in the usual sense. Nurses pointed to the


713


need for better communication and greater professional collaboration as important to their sense of moral agency, however, they acknowledged that even that would not solve all the problems. The fact remained that the suffering was there, and would not go away, and as such, generated moral distress. These results point to the need for increased dialogue within and between medicine and nursing around the ethical aspects of decisions. Pike (1991) provides support for this notion. She found that the incidence of moral outrage among nurses substantially decreased in a hospital unit dedicated to fostering nurse±physician collaboration. Similarly, one might expect that physicians' moral burden would lessen if they shared their uncertainty and distress with nurses. As one nurse suggested, `they don't have to do it by themselves'. Viney's (1996) study led to similar conclusions, namely, that a model of communication supporting and enhancing collaborative, multidisciplinary ethical decision making was needed. Another important implication of the study is the need for administrators to recognize the burden carried by practitioners who are required to witness suffering as part of their daily work (Rodney 1998). Kelly (1998b) suggests that moral decisions are shaped by conditions in the workplace. Unhealthy work environments and poor patient outcomes may result from a failure of doctors and nurses to work together in a collaborative fashion (Larson 1999). Clearly, strategies for supporting decision-making and opportunities for caregivers to engage in moral discourse must be a focus for administrators who wish to sustain a healthy, collaborative work environment (Baggs 1994, Johnson et al. 1995).

Conclusions Although there were some ethics-related differences between doctors and nurses, our principal ®nding was one of similarities. Observed differences appeared to be more a function of the hospital's hierarchical structure and the designated roles of doctors and nurses, rather than a difference in moral commitment or reasoning. There was little evidence of the `care vs. cure' distinction between the two. Instead, it was evident that doctors and nurses are confronted with different problems and ask different questions in the same care situation. In end-of-life decision making it is an inescapable fact that doctors must make the decisions and nurses must live with them. Both responsibilities carry moral obligation and burden. We propose that the moral distress experienced by both groups could be reduced through cross-disciplinary discussion and mutual recognition of the burden carried by the other. 714

Acknowledgements We would like to thank Dr Elizabeth Flagler for her participation earlier phases of this work. This research was funded by Alberta Association of Registered Nurses.

References van der Arend A.J. & Remmers-van den Hurk C.H.M. (1999) Moral problems among Dutch nurses: a survey. Nursing Ethics 6, 468±482. Baggs J. (1994) Development of an instrument to measure collaboration and satisfaction about care decisions. Journal of Advanced Nursing 20, 176±182. Beauchamp T. & Childress J. (1989) Principles of Biomedical Ethics, 3rd edn. Oxford University Press, New York. Bishop A.H. & Scudder J.R. (1996) Nursing Ethics: Therapeutic Caring Presence. Jones & Bartlett, Boston. Cooper M. (1988) Covenantal relationships: grounding for the nursing ethic. Advances in Nursing Science 10, 48±59. Fry S. (1989) Toward a theory of nursing ethics. Advances in Nursing Science 11, 9±22. Gadow S. (1999) Relational narrative: the postmodern turn in nursing ethics. Scholarly Inquiry for Nursing Practice: An International Journal 13, 57±67. Gramelspacher G.P., Howell J.D. & Young M.J. (1986) Perceptions of ethical problems by nurses and doctors. Archives of Internal Medicine 46, 577±578. Grundstein-Amado R. (1992) Differences in ethical decision-making processes among nurses and doctors. Journal of Advanced Nursing 17, 129±137. Grundstein-Amado R. (1993) Ethical decision making processes among nurses and doctors. Journal of Advanced Nursing 18, 1701±1709. Gustafson J. (1990) Moral discourse about medicine: a variety of forms. Journal of Medicine and Philosophy 15, 125±142. Johanson W. (1994) Promoting collaboration in ethical decision making. Critical Care Nurse 14, 96±99. Johnson J.V., Hall E.M., Ford D.E., Mead L.A., Levine D.M., Wang N. & Klag M.J. (1995) The psychosocial work environment of physicians: the impact of demands and resources on job dissatisfaction and psychiatric distress in a longitudinal study of Johns Hopkins Medical School graduates. Journal of Occupational & Environmental Medicine 37, 1151±1159. Joudrey R. & Gough J. (1999) Caring and curing revisited: student nurses' perceptions of nurses' and physicians' ethical stances. Journal of Advanced Nursing 29, 1154±1162. Kelly B. (1998a) Preserving moral integrity: a follow-up study with new graduate nurses. Journal of Advanced Nursing 28, 1134±1145. Kelly C. (1998b) Investing or discounting self: are moral decisions shaped by conditions in the workplace? Advanced Practice Nursing Quarterly 4(2), 8±13. Kuhse H., Singer P., Rickard M., Cannold L. & van Dyk J. (1997) Partial and impartial ethical reasoning in health care professionals. Journal of Medical Ethics 23, 226±232. Larson E. (1999) The impact of physician±nurse interaction on patient care. Holistic Nursing Practice 13, 38±46.


1 Philosophical and ethical issues Lindseth A., Marhaug B., Norberg A. & Uden G. (1994) Registered nurses' and physicians' re¯ections on their narratives about ethically dif®cult care episodes. Journal of Advanced Nursing 20, 245±250. Lipp A. (1998) An enquiry into a combined approach for nursing ethics. Nursing Ethics 5, 122±138. McClure M.L. (1991) Nurse±physician con¯ict. Journal of Professional Nursing 7, 329. Parker R.S. (1990) Measuring nurses' moral judgements. Image: Journal of Nursing Scholarship 22, 213±218. Peter E. & Gallop R. (1994) The ethic of care: a comparison of nursing and medical students. Image: Journal of Nursing Scholarship 26, 47±51. Pike A.W. (1991) Moral outrage and moral discourse in nursephysician collaboration. Journal of Professional Nursing 7, 351± 363. Ray M.A. (1998) A phenomenologic study of the interface of caring and technology in intermediate care: toward a re¯exive ethics for clinical practice. Holistic Nursing Practice 12, 69±77. Robertson D.W. (1996) Ethical theory, ethnography, and differences between doctors and nurses in approaches to patient care. Journal of Medical Ethics 22, 292±299. Rodney P. (1998) Towards ethical decision-making in nursing practice. Canadian Journal of Nursing Administration 11, 34±45. Rodney P. & Starzomski R. (1993) Constraints on the moral agency of nurses. Canadian Nurse 89(9), 23±26. Soderberg A. & Norberg A. (1993) Intensive care: situations of ethical dif®culty. Journal of Advanced Nursing 18, 2008±2014.

Ethical problems in end-of-life decisions Sorlie V., Lindseth A., Uden G. & Norberg A. (2000) Women physicians' narratives about being in ethically dif®cult care situations in paediatrics. Nursing Ethics 7, 47±62. Strauss A. & Corbin J. (1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 2nd edn. Sage, Thousand Oaks. Sundin-Huard D. & Fahy K. (1999) Moral distress, advocacy and burnout: theorising the relationships. International Journal of Nursing Practice 5, 8±13. Taylor S.L. (1995) Quandary at the crossroads: paternalism versus advocacy surrounding end-of-treatment decisions. American Journal of Hospice and Palliative Care 12, 43±46. Twomey J. (1989) Analysis of the claim to distinct nursing ethics: normative and nonnormative approaches. Advances in Nursing Science 17, 1028±1034. Uden G., Norberg A., Lindseth A. & Marhaug V. (1992) Ethical reasoning in nurses' and physicians' stories about care episodes. Journal of Advanced Nursing 17, 1028±1034. Viney C. (1996) A phenomenological study of ethical decisionmaking experiences among senior intensive care nurses and doctors concerning withdrawal of treatment. Nursing in Critical Care 1, 182±187. Walker R., Miles S., Stocking C. & Siegler M. (1991) Physicians' and nurses' perceptions of ethics problems on general medical services. Journal of General Internal Medicine 6, 424±429. Wurzbach M.E. (1999) Acute care nurses' experiences of moral certainty. Journal of Advanced Nursing 30, 287±293. Yarling R.R. & McElmurry B.J. (1986) The moral foundation of nursing. Advances in Nursing Science 8, 63±73.


715